Stock Pulse
StockPulse is a university project that analyzes the sentiment of financial news related to publicly traded companies. It combines a lightweight backend architecture (C# API + FastAPI NLP service) with a responsive React frontend.
Tech Stack
Overview
StockPulse provides real-time sentiment analysis of the latest news for a given stock ticker. It leverages the NewsData.io API for fetching financial news and a local NLP service for sentiment classification.
Users can:
- Search for a stock by ticker symbol or company name.
- View recent related news headlines.
- See the aggregated sentiment (positive, negative, or neutral).
- Access full news articles via external links.
Architecture
Frontend (React + Vite + Tailwind + DaisyUI)
│
▼
C# API (ASP.NET Core) ← loads ticker data & validates symbols
│
▼
Python FastAPI (NLP Service) ← fetches news, performs sentiment analysisProxy setup
Caddy or Vite’s proxy forwards /api/* requests to the backend services, enabling local development without CORS issues.
Features
- Search by ticker symbol with autocomplete.
- Server-side validation of valid tickers.
- Sentiment analysis using NLP.
- Cached responses for performance.
- Responsive, minimalist UI with DaisyUI.
- Error handling for missing or unavailable news.
Tech Stack
| Layer | Technology |
|---|---|
| Frontend | React, Vite, Tailwind CSS, DaisyUI |
| Backend (API) | ASP.NET Core (.NET 8) |
| NLP Service | Python, FastAPI, httpx, Pydantic |
| Data Source | NewsData.io |
| Deployment | Caddy, Docker-ready |
| Language | JavaScript / TypeScript, C#, Python |
Directory Structure
StockPulse/
├── frontend/ # React + Vite client
│ ├── src/
│ │ ├── components/
│ │ ├── hooks/
│ │ └── App.jsx
│ └── vite.config.js
│
├── StockPulse.Api/ # C# backend
│ ├── Controllers/
│ ├── Data/tickers.csv
│ └── Program.cs
│
└── nlp/ # Python NLP microservice
├── routers/
├── services/
└── main.py
Run Locally
git clone https://github.com/Puchungualotsqui/stock_pulse
cd StockPulse
./start_stockpulse.sh devthen open http://localhost:5173/
Enviroment Variables:
| Variable | Description |
|---|---|
NEWS_API_KEY | API key from NewsData.io |
PORT | Optional, overrides default port for the NLP service |
Academic Notice
This project was created as part of a university data science course. It is not intended for commercial or financial use. Data and sentiment classifications are for educational demonstration only.